I would like to have a chat to talk on a personal plan enough to pass the Turing test.

I got to study some natural language processing (NLP) but some that usually hide a lot of mechanical, so I think they are also much more sensitive than words and get phrases more fluid.

The problem is that I do not even have the idea to start. I did some testing with a NLP chatterbot library for python, but I do not know if it's possible to join PLN to Networks or how to use a neural network and I can not even use python or if I have to use a specific language.

In addition, I am using the book Artificial Intelligence - Russell and Norvig as the basis for the article and I would like recommend me others.

  • 1
    $\begingroup$ Could you explain what "a chat to talk on a personal plan" means? Is it some constraint on the Chatbot, so that it only has to be convincing when discussing "a personal plan"? If so, what is "a personal plan". If not, then please explain more what you mean by that phrase. $\endgroup$ – Neil Slater Sep 14 '18 at 15:16
  • $\begingroup$ I'm sorry if I misunderstood. I meant that this chatbot is for deep personal conversations and so it can not be mechanical like a chatbot made by a simple search tree. $\endgroup$ – Mariana Bayonetta Sep 14 '18 at 17:13
  • $\begingroup$ Sounds great for you to work on chatbot project.However,can you rephrase your question clear and your will get rightful answer! $\endgroup$ – quintumnia Sep 14 '18 at 17:38

I would not recommend using neural networks and NLP together to create a system sufficiently capable of conversation/dialogue that it would pass that current crop of Turing-like tests.

Conversations follow certain rules and regularities (which we have only partially discovered so far), and training an ANN with dialogues in order to pick up those regularities is simply not feasible. In conversations you have a memory of what has been mentioned previously, you build up assumptions about the intentions of your dialogue partner, and keep track of the current topic and sub-topics. This is far too complex to be reduced to a machine learning approach.

As a starting point I would suggest looking at ELIZA, developed by Weizenbaum in the mid-1960s. There are plenty of implementations in various programming languages available. Use that as a starting point to extend the capabilities according to topics you want to talk about, and store in memory what the user has said before, trying to refer back to it, etc. This is a lot easier to do with 'symbolic' AI rather than subsymbolic processing.

A lot of current tech companies offer chatbot variants based on machine learning, but they rarely go beyond intent recognition or simple question-answer dialogues. For more sophisticated dialogues they are simply not suitable.

(Disclaimer: I work for a company producing conversational software)

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.